* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download The nucleotide sequence and derived amino acid
Catalytic triad wikipedia , lookup
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
Two-hybrid screening wikipedia , lookup
Restriction enzyme wikipedia , lookup
Agarose gel electrophoresis wikipedia , lookup
Endogenous retrovirus wikipedia , lookup
Gel electrophoresis of nucleic acids wikipedia , lookup
Multilocus sequence typing wikipedia , lookup
Bisulfite sequencing wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Messenger RNA wikipedia , lookup
Epitranscriptome wikipedia , lookup
Non-coding DNA wikipedia , lookup
Genomic library wikipedia , lookup
Ancestral sequence reconstruction wikipedia , lookup
Nucleic acid analogue wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Homology modeling wikipedia , lookup
Protein structure prediction wikipedia , lookup
Biochemistry wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Community fingerprinting wikipedia , lookup
Metalloprotein wikipedia , lookup
Point mutation wikipedia , lookup
Gene, 25 (1983) 325-332 325 Elsevier GENE 884 The nucleotide sequence and derived amino acid sequence of eDNA coding for mouse carbonic anhydrase II (Comparison with mammalian carbonic anhydrase isozymes; evolution; unique residues; long 3' noneoding region; recombinant DNA) Peter J. Curtis *, Elizabeth Withers, Donald Demuth, Rosemary Watt, Patrick J. Venta **, and Richard E. Tashian ** The Wistar Institute of Anatomy and Biology. 36th Street at Spruce, Philadelphia, PA 19104 (U.S.A.) Tel. (215) 898-3859, and ** Department of Human Genetics, University of Michigan Medical School, Ann Arbor, M148109 (U.S.A.) Tel. (313) 764-1359 (Received May 19th, 1983) (Revision received July 28th, 1983) (Accepted August 3rd, 1983) SUMMARY The nucleotide sequence of a clone containing mouse carbonic anhydrase (CA) eDNA in pBR322 has been determined. The cloned eDNA contains all of the coding region except for nueleotides specifying the first eight amino acids, and all of the 3' noncoding region, which consists of 700 nucleotides. A eDNA clone was identified which contains an additional 54 bp at the 5' end, so that the complete amino acid sequence of mouse CA could be deduced. This sequence showed a 73-81% homology with other mammalian CA form II isozymes, 56--63 with form I isozymes, and 52-56 ~o with form III isozymes. By examination of the amino acids which are unique and invariant for each isozyme, the mouse amino acid sequence was found to contain 16 of the 23 residues that are unique and invariant to mammalian CA form II isozymes, but only one or no residue for forms I and III, respectively. INTRODUCTION The carbonic anhydrases (EC 4.2.1.1) are zinc metalloenzymes of apparently great evolutionary antiquity which catalyze the reversible hydration of CO2 (for recent reviews see Pocker and Sarkanen, 1978; Lindskog, 1982; Tashian et al., 1983). At least three isozymes of CA, termed CA I, CA II and * To whom correspondence and reprint requests should be sent at the first address. Abbreviations: bp, base pairs; CA, carbonic anhydrase. 0378-1119/83/$03.00 © 1983 Elsevier Science Publishers CA III, appear to be characteristically present in amniotes (reptiles, birds and mammals) where they are under the control of separate genetic loci. The CA II genes are expressed in a wide variety of cells and tissues, whereas the CA I and CA III genes seem to have a more limited tissue expression (Tashian, 1977; Tashian et al., 1983). The complete, or partially (>60~o) complete, amino acid sequences of six CA I, seven CA II and two CA III isozymes have now been determined from 11 species of mammals (Tashian et al., 1983). These are all monomeric enzymes 259 or 260 amino acid residues in length (Mr approx. 29000). When 326 the sequences of all of these CA isozymes are compared, about 30~ of the residues were found to be identical. Despite this relatively high degree of evolutionary homology, however, the often disparate physicochemical and kinetic properties, and considerable variation in tissue distribution of the CA isozymes, suggest different physiological roles for each isozyme. Although some limited sequence data have been obtained for the CA III isozyme of the rat (Carter et al., 1981), and what appears to be CA II cDNA from the mouse (Curtis, 1983), no extensive nucleotide or amino acid sequences are as yet available for a CA isozyme from any rodent species. In this respect, it will be very useful to determine the sequences of the CA isozyme genes of mice, not only because of their obvious relevance to evolutionary and genetic studies, but also because of their potential use in developmental studies where it is important to identify the CA isozymes, e.g., during normal red cell development, or in transformed mouse erythroleukemia cells (Curtis, 1983). In the present report, the complete nueleotide sequence which includes the entire coding region, of mouse CA cDNA cloned from the mRNA purified from mouse anemic spleen has been determined, and identified as CA II by comparison with the known sequences of other mammalian CA isozymes. MATERIALS AND METHODS (a) Isolation of plasmid clones The characterization of a clone which contained CA mRNA derived from Balb/c mice, p6-69, has been described (Curtis, 1983) and shall be designated as pMCAII. Plasmid was isolated from Escherichia coli containing pMCAII by the procedure of IshHorowitz and Burke (1981). Additional plasmids containing CA sequences were identified in a plasmid library containing eDNA sequences from mouse anemic spleen. To make the plasmid eDNA library, double-stranded DNA was synthesized from poly(A) + RNA of mouse anemic spleen, treated with S 1 nuclease, tailed with dCMP residues and selected for a size greater than 1000 bp (Curtis, 1983). The tailed DNA was annealed with dG-tailed pBR322 and the hybrid DNA used to transfect E. co/i HB101. One million tetracyclineresistant colonies were collected from the agar plates and plasmid extracted as described above. The pooled plasmid DNA was used to transfect E. coli. To screen the colonies, pMCAII was cut with PstI and fragments derived from the cloned sequences were purified by agarose gel electrophoresis and labeled using calf thymus primers and E. coli DNA polymerase I (Summers, 1975). (b) DNA sequencing Restriction fragments derived from pMCAII by digestion with PstI, BamHI and HindlII, were subcloned into M13mp8 and sequenced by the dideoxy chain termination method (Sanger etal., 1977; Messing and Vieira, 1982). Additional sequences were determined by the procedure of Maxam and Gilbert (1977) using 3' end-labeled restriction fragments. RESULTS AND DISCUSSION (a) Strategy of DNA sequencing The plasmid pMCAII for mouse CA contains an insert of about 1500 bp with two internal PstI sites resulting in three fragments of 700, 450, 320 bp (Fig. 1). These fragments were cloned into M13mp8 cut with Psi I. The resultant six clones were sequenced by the dideoxy chain termination method. The two smaller fragments were sequenced completely. For the larger fragment, the middle section of the sequence was obtained from M13mp8 clones containing the BamHI-BglII fragment as well as the HindIII-PstI. Additional segments were sequenced by the Maxam and Gilbert method (1977). After digestion separately with BamHI and HindlII, the 3' ends were labeled, and singly labeled fragments produced by further digestion with restriction enzymes were purified by agarose gel electrophoresis. The sequence derived from pMCAII could be translated into an amino acid sequence that shared extensive homology with the other mammalian CA II isozyme sequences. When these amino acid sequences were aligned, the sequence for the first 327 5' end of mRNA L- It t t L! it ........... i,, pBR322 II qt, u I I 0 I I I I I I 0.5 I i 1.0 I I I I 1.5 Kb Fig. 1. Restriction map of pMCAII. The arrows indicate the direction and extent of the sequence that was determined by the chain termination method and by the Maxam and Gilbert (]977) procedure. eight amino acids from the N terminus was missing. In order to obtain this sequence, a library of mouse cDNA sequences derived from poly(A)+RNA of anemic spleen was screened using the inserted sequences of pMCAII as a labeled probe. Approx. 100 colonies were identified, of which 20 were selected with the strongest response. Plasmid was prepared from these colonies, digested with Pst I and 1 2 3 bp - 700 - 450 -- 320 Fig.2. Agarose gel electrophoresis of plasmids containing mouse C A eDNA. Plasmids were digestedwith Psi I and electrophoresed in a 1.5% agarose gel. Lanes I, 3: two plasmids containing C A II eDNA. Lane 2: pMCAII. analyzed by agarose gel electrophoresis (Fig, 2). From the sequencing it had been determined that the 320-bp fragment contained sequences of the 5' end of the mRNA, the 450-bp fragment contained sees quences of the 3' end of the coding region of the mRNA, while the 700-bp fragment contained almost all of the 3' noncoding region. Two plasmids were identified which contained the 450-bp fragment, but had a fragment of approx. 350-400 bp instead of a 320-bp fragment. They also had a third fragment somewhat smaller than the 700-bp fragment of pMCAII. Presumably these two plasmids contain extra sequences at the 5' end, but lack sequences from the 3' end. The shortened 3' end probably arose from incomplete synthesis of the double,stranded eDNA. Only one RNA species was detected by hybridization ofpMCAII to RNA fractionated by gel electrophoresis and transferred to aminophenylthioether paper (not shown). In addition the size of the 3' end fragment was variable in several different plasmids and there is only one poly(A) addition site in the 3' noncoding sequence. The two plasmids were digested with PstI and Bali and the digest ligated to M13mp8 digested with PstI and Sinai. The appropriate clones of M 13 were identified by a hybridization test with a M13 clone containing the 320-bp fragment of pMCAII. The M13 clones derived from the two plasmids were sequenced as above. The combined nuclootidc sequences are shown in 328 3O CTGCTCTGCCCAATCACCGGCGTGACCATG MET 120 TCC CAC CAC TGG GGA TAC AGC AAG CAC AAC GGA CCA GAG AAC TGG CAC AAG GAC TTC CCC ATT GCC AAT GGA GAC CGG CAG TCC CCT GTG Set His Hts Trp Gly Tyr Set Lys His Asn Gly Pro Glu As.._~nTrp His Lys Asp Phe Pro l i e Aia Asn Gly Asp Ar9 Gln Set Pro Val 30 1 15 210 GAT ATT GAC ACA GCA ACT GCC CAC CAT GAC CCT GCC CTA CAG CCT CTG CTC ATA TCT TAT GAT AAA GCT GCG TCC AAG AGC ATT GTC AAC Asp I l e Asp Thr Ala Thr Ala Hts His Asp Pro Ala Leu Gl_.~nPro Leu Leu l i e Ser Tyr Asp Lys Ala Ala Set Lys Set l i e Val Ash 60 45 300 AAC GGC CAC TCC TTT AAC GTT GAG TTT GAT GAC TCT CAG GAC AAT GCA GTG CTG AAA GGA GGA CCC CTC AGT GAC TCC TAC AGA TTG ATC Asn Gly Hts Set Phe Asn Val GIu Phe Asp Asp Set Gln Asp Asn Ala Val Leu Lys Gly Gly Pro Leu Ser Asp Ser Tyr Ar 9 Leu l l e --7~ 90 390 CAG TTT CAC TTT CAC TGG GGC TCA TCT GAT GGC CAG GGC TCT GAG CAC ACT GTG AAC AAA AAA AAA TAT GCT GCA GAG CTT CAC TTG GTT Gin Phe Hts Phe His TPp Gly $er Set Asp Gly Gin Gly Ser Glu His Thr Val Asn Lys Lys Lys Tyr Ala Ala Glu Leu His Leu Val 105 120 480 CAC TGG AAC ACC AAA TAT GGG GAC TTT GGA AAA GCT GTG CAG CAA CCG GAT GGA TTG GCT GTT TTG GGC TAT TTT TTG AAG ATT GGA CCT HfS Trp Asn Thr kys Tyr Gly Asp ~he Gly kys Ala Val Gin Gln Pro Asp Gly Leu Ala Val Leu Gly Tyr Phe Leu Lys l i e Gly Pro 135 T6~ fi/O GCC TCA CAA GGC CTT CAG AAA GTC CTT GAA GCA CTG CAT TCC ATT AAA ACA AAG GGG AAG CGT GCG GCC TTT GCT AAC TTC GAC CCT TGC Ala Set Gln Gly Leu Gln Lys Val Leu Gl__~uAla Leu His Ser [ l e Lys Thr Lys Gly Lys Ar9 Ala Ala Phe Ala Asn Phe Asp Pro Cys 165 ~ -660 TCC CTT CTT CCT GGA AAC TTG GAC TAC TGG ACA TAC CCT GGC TCT CTG ACC ACT CCG CCT CTG CTG GAP, TGT GTG ACC TGG ATC GTG CTC Ser Leu Leu Pro Gly As.._nnLeu Asp Tyr Trp Thr Tyr Pro Gly $er Leu Thr Thr Pro Pro Leu Leu G|u Cys Val Thr Trp l i e Va| |eu 195 210 75O AGG GAG CCC ATT ACT GTC AGC AGC GAG CAG ATG TCT CAT TTC CGT ACG CTG AAC TTC AAT GAG GAG GGG GAT GCT GAA GAA GCG ATG GTG Ar9 Glu Pro l i e Thr Val Set Set Glu Gin Her Ser His Phe Ar9 Thr Leu Asn Phe Asn Glu Glu GJy Asp Ala G|u Glu Ala Her Val 225 -~ 240 810 840 GAC AAC TGG CGT CCA GCT CAG CCG CTA AAG AAT AGA AAG ATC AAA GCG TCC TTT AAG TAAAACAACCCTGCAGCAGGGGTCCGAAGGCACAAGTGT Asp Asn Trp At9 Pro Ala Gln Pro keu Lys ASh Ar9 Lys l l e kys Ala Set Phe Lys 255 880 930 GAcCGCCTCTCTGTAGCTAAGCGCAGTTA•GG•TGGGTGATTTGGATC•CGACT•GcAT•TGGTATTGTAGAC•TTTTAC•TCT•ATCCGTTGTG•TTACTAA•AAAAG•AAGA••CAGG 960 1000 1050 1080 TGTCTCATGTGGTGGCAGCACAGTGGCAGGCCAGTGGTCAACTTAGGGCATCTTTTCTCTGCCACGGCAG~GCAATGCAAAGAGCAGA~ATGGCCTCTTG~TTCTCTTCACAGCCATAG G 1120 I170 1200 ATAATGAA•A•TCAGG••TGHTGTTAAAATG•TATTTTAAAA••A•A•GAAGGTAGGATAATTAATTACAAGT•CA•AT•ATGAGA•AAACTGAAGTAACTT•GGCAAAACAGGTAAAC 1240 1290 1320 AGT•ATAGTHT•TGATTATAAATGAG•T•AATGTTCA•C•TTC•AAGA••TTATATTAAAGAAAAAA•TTTAAAAAGCTTATATATTTGTAG•AAAGTTATT•TTAAATA•GAATTATG 1360 1410 1440 TTATAA~TTAGTGA~TTTTGATTT~TAGAGGTGTAAATGAGGA~GTAAAAATTGATATAGTTGTGA~A~AGAGTATATTTCC~TTC~GATAA~ATA~CA~AACACAATGG~AATGTAT 1480 TTTAGATATATTCTCTAATAAAATTGAGAACTCT Fig. 3. Nucleotide sequence of cloned mouse CA II eDNA and the deduced amino acid sequence. The underlined amino acid residues are unique to the mouseCAII sequence as compared to the CAII sequence ofhuman, sheep, ox, rabbit and horse. Fig. 3. There are 27 nucleotides preceding the initiation codon in the 5' noncodingregion: No homology was found in the 5' noncoding region with globin mRNA. Examination of the coding region shows that the use of triplets does. not differ significantly from the distribution found in globin mRNA (Kafatos et al., 1977; Heindell et al., 1978; Curtis, 1980), except that in CA II the triplet C U G for leucine is not used as extensively as in giobin, while for glycine, G G A is used frequently and G G U is not used at all. However, since CA. II constitutes < 1 of total protein in most mammalian erythrocytes as well as being present at low levels in almost all tissues (Tashian, 1977), it is less likely to be affected by the 329 relative concentrations of tRNAs which in reticulocytes correlates with the amino acid composition of globin (Smith, 1975). The growing number of processed transcripts characterized by long ( > 500 bp) noncoding regions (e.g., dihydrofolate reductase mRNA, porcine fl-neo-endorphin mRNA, seal myoglobin gene, human plasminogen activator mRNA, and human hypoxanthine phosphoribosyltransferase mRNA; Setzer et al., 1980; Kakidani et al., 1982; Blancetot etal., 1983; Penniea etal., 1983; Jolly et al., 1983) suggests that such a phenomenon may not be as uncommon as previously thought. However, no significance has as yet been associated with large 3'-nontranslated regions. In CA II mRNA following the termination signal the same reading frame remains open for 324 nucleotides before several stop codons are encountered. It would be of interest to know if read-through of the normal termination signal occurs at all. Located 18 nucleotides from the beginning of the poly(A) region at the 3' end is the sequence AATAAA which is found in all poly(A) ÷-containing mRNAs and acts as a signal either for cleavage of the pre-mRNA or as a recognition site for polyadenylation. (b) Identification o f the m o u s e C A c D N A as a C A II isozyme From the nucleotide sequence, the complete amino acid sequence of a mouse CA was deduced. One of the most effective methods that can be used to assign an unknown CA to a specific isozyme type (i.e., CA I, CA II or CA III) is to compare the residues in its sequence to those residues which are invariant in all examples of that isozyme sequenced to date, but TABLE II Comparison of mouse CA II amino acid sequence with amino acid sequences from selected mammalian CA isozymes Species a CA isozymes CA I Human Rabbit Ox Horse CA II CA III Residues identical to mouse CA II (~) 63 81 56 56 77 - 58 58 75 78 52 - a See footnote to Table III for references. have a different residue at this position in the other two isozymes. For example, all ofthe CA II isozymes have His residues at positions 2 and 3 (as does the mouse CA isozymv), whereas the His does not appear at these positions in any of the CA I or CA III isozymes. This information is summarized in Table I. As can be seen, the mouse CA sequence shows only one residue in common with the 21 residues that are unique and invariant in the CA I isozymes, and no residues in common with the 31 residues unique to the CA III isozymes. However, the fact that 16 of the 23 residues that were considered previously to be unique and invariant to the CA H isozymes were identical to the mouse sequence, strongly suggests that the mouse CA sequence is most closely related to the CA II isozymes ofmammals. The fact that the mouse CA II has different amino acids in seven positions, which are invariant in all other CA IIs, may be related to the unusual properties of mouse CA II, as discussed below. TABLE I Comparison of the putative mouse CA II amino acid sequencewith the unique and invariant residues of mammalian CA I, CA II and CA III isozymes CA isozyme Number of species sequenced a Number of unique and invariant residues CA I 6 21 CA II 7 23 b CA III 2 31 Number of mouse "CA IF' residues identical to the unique and invariant residues of Other CA isozymes (~) 1 (5) 16 (70) 0 (0) a For referencesto these CA isozymesequences,see Tashian et al. (1983) and citations therein. b These are at positions 2, 3, 7, 18, 26, 55, 57, 66, 68, 75, 86, III, 124, 127, 128, 130, 136, 164, 171,173, 228, 229, 253 in Fig:3. 330 (c) Evolutionary considerations ¢- The fact that all three CA isozymes are probably present in all species of amniotes indicates that the duplications which gave rise to the CA isozyme genes occurred more than 300 million years ago (Tashian et al., 1983). Over this evolutionary period, it appears that the CA I and CA II genes have remained closely linked since they have been assigned to a region close to the centromere of chromosome 3 in the mouse (Eicher et al., 1976), and in all likelihood, are also linked on chromosome 8 in humans (Venta et al., 1983). Pairwise comparisons of the mouse CA II sequence with the sequences of the other CA isozymes (Table II) provide an approximation of the evolutionary distances among these isozymes. As can be seen in Table II, the mouse CA II shows a 73-81~o homology with the CA II isozymes, a 56--63% homology with the CA I isozymes, and a 52 and 56~ homology with the CA III isozymes. 8 U E _ cad o . o,) ¢- L., o o~ o (d) Residues unique to the mouse CA II sequence The following discussion of the residue positions of the mouse CA II isozyme is based on the assumption that its three-dimensional structure is similar to that of human CA II. For references to the highresolution molecular structure of human CA II, see Vaara (1974) and Notstrand et al. (1975). When the mouse CA II sequence is compared with the complete CA II sequences of human, sheep, ox, rabbit and horse (for references, see footnote to Table III), 33 residues were found to be unique to the mouse sequence and these are underlined in Fig. 3. All but five of the 33 residues are located on the surface of the human CA II molecule. It is of interest that 15 of the 33 amino acids are invariant in the non-mouse sequences. These residues which are listed in Table III, may contribute to the somewhat unusual properties that have been observed for the mouse CA II isozyme. These properties include the relatively rapid alterations (probably as a consequence of aging) in their high-pressure liquid chromatography elution patterns and electrophoretic patterns comp"ared to the more stable patterns of the CA II isozymes of other mammals (D. Hewett-Emmett and Y.-S.L. Yu, unpublished observations). With the possible exception of Gin-153 which may have ,.-C. < -i 0 ¢~ O "i 8 .E d~ m 8~ 8 E ," E i. o~ e~ 331 the effect of extending c~-helixE at its N-terminus, no obvious changes in secondary structures, aromatic clusters and hydrophobic core residues would appear to result from the amino acid differences at these 15 positions. Of the remaining 18 residues unique to mouse CA II but variable in the CA II isozymes of other mammals, two residues (Tyr- 144 and Cys-180) are of interest. The residues at position 144 in the other CA II isozymes are Ile, Leu and Val. Residue 144 is located in the middle of fl-structure segment 6 and has been assigned to the hydrophobic core of the human CA II molecule. The presence of Tyr at this position (since it is less hydrophobic than Ile, Val and Leu) might affect the stability of the mouse CA II molecule. The residue at position 180 is Arg, Ser or Gly in the other CA II isozymes and is located in 3~o-helix F of human CA II. Since this residue is found on the surface of the human CA II molecule, it is possible that Cys-180 in mouse CA II may form intermolecular dimers with other CA II molecules. ACKNOWLEDGEMENTS We thank Dr. G. Rovera and Dr. David HewettEmmett for their helpful comments during the preparation of the manuscript. This work was supported in part by U.S. Public Health Service Grant GM-24681 (R.T.) and National Institutes of Health Grants CA-30136, CA-10815 and CA-25875 (P.J.C.). REFERENCES Blanchetot, A., Wilson, V., Wood, D. and Jeffrey, A.J.: The seal myoglobin gene: an unusually long globin gene. Nature 301 (1983) 732-734. Carter, N.D., Hewett-Emmett, D., Jeffery, S. and Tashian, R.E.: Testosterone-induced, sulfonamide-resistant carbonic anhydrase of rat liver is indistinguishable from skeletal muscle carbonic anhydrase III. FEBS Lett. 128 (1981) 114-118. Curtis, PJ.: Globin mRNA in friend cells: its structure, function and synthesis. Bioehim. Biophys. Acta 605 (1980) 347-364. Curtis, P.J.: Cloning of mouse carbonic anhydrase mRNA and its induction in mouse erythroleukemia cells. J. Biol. Chem. 258 (1983) 4459--4463. Eicher, E.M., Stern, R.H., Womack, J.E., Davisson, M.-T., Roderick, T.H. and Reynolds, S.C.: Evolution of mammalian carbonic anhydrase loci by tandem duplication; close linkage of Car-I and Car-2 to the centromere region of chromosome 3 of the mouse. Bioehem. Genet. 14 (1976) 651-660. Ferrell, R.E., Stroup, S.K., Tanis, R.J. and Tashian, R.E.: Amino acid sequence of rabbit carbonic anhydrase II. Biochim. Biophys. Acta 533 (1978) I-I1. Heindell, H.C., Liu, A., Paddock, G.V., Studnicka, G.M. and Salser, W.A.: The primary sequence of rabbit ~-giobin mRNA. Cell 15 (1978) 43-54. Henderson, L.E., Henriksson, D. and Nyman, P.O.: Primary structure of human carbonic anhydrase C. J. Biol. Chem. 251 (1976) 5457-5463. [sh-Horowicz, D. and Burke, J.F.: Rapid and efficient cosmid cloning. Nucl. Acids Res. 9 (1981) 2989-2998. Jabusch, J.R. and Deutseh, H.F.: The sequence of the high activity equine erythrocyte carbonic anhydrase. Fed. Prec. 40 (1981) 1677. Jolly, D.J., Okayama, H:, Berg, P., Esty, A.C., Filpula, D., Bohlan, P., Johnson, G.G., Shively, J.E., Hunkapillar, T. and Friedmann, T.: Isolation and characterization of a full-length expressible eDNA for human hypoxanthine phosphoribosyltransferase. Prec. Natl. Acad. Sci. USA 80 (1983) 477-481. Kafatos, F.C., Efstratiadis, A., Forget, B.G. and Weissman, S.M.: Molecular evolution of human and rabbit fl-giobin mRNAs. Prec. Natl. Acad. Sci. USA 74 (1977) 5618-5622. Kakidani, H., Furutani, Y., Takahashi, H., Noda, M., Morioto, Y., Hirose, T., Asai, M., Inayama, S., Nakanishi, S. and Numa, S.: Cloning and sequence analysis of cDNA for porcine ~neo-endorphin/dynorphin precursor. Nature 298 (1982) 245-249. Lindskog, S.: Carbonic anhydrase, in Eichhom, G.L. and Marzilli, L.G. (Eds.), Advances in Inorganic Biochemistry, Vol. 4. Elsevier/North-Holland, Ne.w York, 1982, pp. 115-170. Maxam, A.M. and Gilbert, W.: A new method for sequencing DNA. Prec. Natl. Acad. Sci. USA 74 (1977) 560-564. Messing, J. and Vieira, J.: A new pair elM 13 vectors for selecting either DNA strand of double-digest restriction fragments. Gene 19 (1982) 269-276. Norstrand, B., Vaara, I. and Kannan, K.K.: Structural relationship of human erythrocyte carbonic anhydrases B and C, in Markert, C.L (Ed.), Isozymes: Molecular Structure, Vol. 1. Academic Press, New York, 1975, pp. 575-599. Pennica, D., Holmes, W.E., Kohr, W.J., Harkins, R.N., Vehar, G.A., Ward, C.A., Bennett, W.F., Yelverton, E., Seeburg, P.H., Heyneker, H.C. and Goeddel, D.V.: Cloning and expression of human •sue-type plasminogen activator cDNA in E. coll. Nature 301 (1983) 214-221. Pecker, Y. and Sarkanen, S.L.: Carbonic anhydrase: structure, catalytic versatility and inhibition. Adv. Enzymol. 47 (1978) 149-274. Sanger, F., Nicklen, S. and Coulson, A.R.: DNA sequencing with chain-terminatinginhibitors. Prec. Natl. Acad. Sci. USA 74 (1977) 5463-5467. Sciaky, M., Limozin, N., Filippi-Foveau, D.,Gulian, J.-M. and Laurent-Tabusse, G.: Structure primaire de ranhydrase carbonique 6rythrocytalre bovine CI, II. S~quance eompl&e. Biochimie 58(9) (1976) 1071-1082. 332 Setzer, D.R., McGrogan, M., Nanberg, J.H. and Schimke, R.T.: Size heterogeneity in the 3' end of dihydrofulate reductase messenger RNAs in mouse cells. Cell 22 (1980) 361-370. Smith, D.W.E.: Reticulocyte transfer RNA and hemoglobin synthesis. Science (1975) 529-533. Summers, J.: Physical map ofpolyoma DNA fragments produced by cleavage with a restriction enzyme from Haemophilus aegyptius endonuclease R. HaeIII. J. Virol. 15 (1975) 946-953. Tanis, R.J.0 Ferrell, R.E. and Tashian, R.E.: Amino acid sequence of sheep carbonic anhydrase C. Biochim. Biophys. Acta 371 (1974) 534--548. Tashian, R.E.: Evolution and regulation of the carbonic anhydrase isozymes, in Rattazzi, M.C., Scandalios, J.G. and Whitt, G.S. (Eds.), Isozymes, Current Topics in Biological and Medical Research, Vol. 2. Liss, New York, 1977, pp. 21-62. Tashian, R.E., Hewett-Emmett, D. and Goodman, M.: On the evolution and genetics of carbonic anhydrase I, II, and III, in Rattazzi, M.C., Scandalios, J.G. and Whitt, G.S. (Eds.), lsozymes, Current Topics in Biological and Medical Research, Vol. 7. Liss, New York, 1983, pp. 79-100. Tashian, R.E., Hewett-Emmett, D., Stroup, S.K., Goodman, M. and Yu, Y.-S.L.: Evolution of structure and function in the carbonic anhydrase isozymes of mammals, in Bauer, C., Gros, G. and Bartels, H. (Eds.), Biophysics and Physiology of Carbon Dioxide. Springer-Verlag, Berlin, 1980, pp. 165-176. Vaara, I.: The molecular structure of human carbonic anhydrase, form C and inhibitor complexes. Ph.D. Dissertation, University of Uppsala, Sweden, 1974. Venta, P.J., Shows, T.B., Curtis, P.J. and Tashian, R.E.: The polymorphic gene for human carbonic anhydrase II: a new molecular disease marker located on chromosome 8. Prec. Natl. Acad. Sci. USA 80 (1983) 4437-~.A.A.0. Communicated by A.-M. Skalka.